Tracking time

Just a few notes on tracking your time.

• No one is being accused of dishonesty. We're trying to get a handle on how we spend our time, so we can be more efficient. Some of you are naturally more efficient than others. We find that tracking how time is spent, in addition to giving us valuable information, helps each individual focus on how she/he is spending time. Measurement alone should help us improve.
• The spreadsheets I've received so far look like it was easier for you this time than last. Good. Go ahead and reread the instructions, just to make sure we're all on the same page.
• I am comparing the time tracking spreadsheets with your census logs, so please track time, as appropriate, on both.
• On your census logs, track date and start time, date and stop time and exact total time for each soldier.
• DO NOT include anything that isn't inputting or soldier related on your census log.
• Track how you spend ALL your time (not just searching and inputting) on your time tracking spread sheet.
• The time tracking spreadsheet  totals for each day should match what you submit on your time sheet. If they don't match, it makes it harder for me to evaluate what is happening.
• Some of you have asked if you should track your lunch on the spreadsheet. Initially, I said it didn't matter. I've changed my mind. Please don't track it on your spreadsheet. It's confusing me.

Let me know if you have feedback.

Husband or head of household?

Recently, there have been several instances in the 1880 census where the enumerator writes "husband" instead of "head" for the head of household. We've decided to make a rule that when this happens, the relationship should be "1" not "7."

Note: If the woman is ever listed as "head," and the husband is listed as "husband" on the manuscripts, then mark the husband "7" in the screens.

May 2016 checking stats

In May, we checked 44 soldiers under our system. I've reviewed all of the checks, and I've tallied the number of differences. Some of these are errors, and some are judgment calls. Here are the categories and the total number of differences for each category. 

GRID Errors

• MILIN?/MAR? - 4
• Missing HH member - 2
• Duplicate people - 1
• Wrong person - 0
• Other - 2

Inferred Relationships

• Incorrect relationships - 14

Census Errors

• Name - 40!
• Typo/Reading/Wrong - 103
• State Code - 0
• Missing/Wrong URL - 1
• Missing data - 46!
• Additional finds - 27
• Quality Code - 17

Death Errors

• Typo/Reading/Wrong - 10
• Missing data - 20
• Missing/Wrong URL/Source - 18
• Quality Code - 2
• Additional finds - 11

Tree Errors

• Missing/Incorrect information/relationships - 18

The total number of differences for all 44 soldiers is 336. This is 31 more differences than last month. Some differences/errors affect the data more than others. If we checked other pensions, we'd probably find similar differences.

Updating birth and death places on the family tree to generate hints

Donna suggested the following tip to her checkee to help generate better hints on the family tree.

For the place of birth and death of the soldier, I like to input a place Ancestry recognizes. I find this helps Ancestry generate better hints. It doesn’t always help though, and it isn’t required.

For example, the mil info gives place of birth as Telmurker, NW. Ancestry doesn’t recognize this place though because Telmurker is a misspelling and NW is a generated code. Since Norway is so unusual, I would change it to only Norway without the city name.

Mil info gives place of death as Rochester, Olmista, MN. Ancestry doesn’t recognize this because Olmista is a misspelling of Olmstead. To figure out Olmista was Olmstead, I started deleting what had been input until Ancestry gave me an option in the drop down box. This doesn’t always work though because the places are too misspelled or they don't exist anymore. In those cases, I just put in the state.

Relationship codes and HH0

Here is something to keep in mind when you're checking household 0. When much of this information was originally collected, we didn't have all the Relationship codes that we have now.

As always, when you're checking old work, please bring the data up to our current standard.

For example, you might find that siblings are marked as Relationship code 4. You should change this to the more accurate, Relationship code 10.

If you've noticed similar issues while checking, please let your fellow inputters know by leaving a comment.

Home value

There has been some confusion when reading home values and determining whether the zeroes are cents or dollars.

As a general rule, the rent will be a two digit number and can be followed by small zeroes for cents (14.00).  Value of a home that is owned will more likely be a four digit number (1400).

Death and the public family tree

A question came recently about using public family trees as death sources. Here is Sandy's answer.

If you find a DOCUMENT (like the photo of the death certificate) on a public tree, it can be counted as a DEATH CERTIFICATE and given a QC1 (assuming it's your guy with proof of parents' names and birth date just like any other death certificate).

A document on a public tree is not the same as an unsourced date written in a public tree. A document is a document and should be treated as such.

Do not take PLACE of death from a tree when you got the DATE of death elsewhere (example Find A Grave). If you found a good death date somewhere other than a tree, then use that (QC3 or higher). If the only place you find the PLACE OF DEATH to go along with that death date is a public tree, then we don't want you to use the place of death, because public trees are last resorts only and are QC4. If each field had a QC, then we probably would allow adding bits of information from a public tree.  But we can't do that.

If you only find the place of death and the death date on the public tree, then you may use the information from the public tree as a QC4.

In a nutshell: a public tree can be used for death information if, and only if, it is the ONLY information you can find. Give it a QC4.  If the tree has a photo of a document as a source of death information, treat that document like any other document with the appropriate QC up to a QC1.